Please use this identifier to cite or link to this item:
https://open.uns.ac.rs/handle/123456789/20129
Title: | A Unification and Generalization of Exact Distributed First-Order Methods | Authors: | Jakovetić Dušan | Issue Date: | 2019 | Journal: | IEEE Transactions on Signal and Information Processing over Networks | Abstract: | © 2018 IEEE. Recently, there has been significant progress in the development of distributed first-order methods. In particular, Shi et al. (2015) on the one hand and Qu and Li (2017) and Nedic et al. (2016) on the other hand propose two different types of methods that are designed from very different perspectives. They achieve both exact and linear convergence when a constant step size is used - a favorable feature that was not achievable by most prior methods. In this paper, we unify, generalize, and improve convergence speed of the methods by Shi et al. (2015), Qu and Li (2017), and Nedic et al. (2016), when the underlying network is static and undirected. We first carry out a unifying primal-dual analysis that sheds light on how these methods compare. The analysis reveals that a major difference between the methods is on how the primal error affects the dual error along iterations. We, then, capitalize on the insights from the analysis to derive a novel method that can reduce the negative effect of the primal error on the dual error. We establish for the proposed generalized method global R-linear convergence rate under strongly convex costs with Lipschitz continuous gradients. | URI: | https://open.uns.ac.rs/handle/123456789/20129 | ISSN: | 2373-776X | DOI: | 10.1109/TSIPN.2018.2846183 |
Appears in Collections: | PMF Publikacije/Publications |
Show full item record
SCOPUSTM
Citations
79
checked on May 10, 2024
Page view(s)
31
Last Week
1
1
Last month
2
2
checked on May 10, 2024
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.